Open-ended Computer Science Problem Solving on Frontier-CS
61.33Mean ScoreAdaEvolve
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| AdaEvolveBackbone=GPT-5, Number of LLM calls=502026.02 | 61.33 | 75.15 | |
| OpenEvolveBackbone=GPT-5, Number of LLM calls=502026.02 | 50.75 | 56.37 | |
| ShinkaEvolveBackbone=GPT-5, Number of LLM calls=502026.02 | 47.79 | 46.22 | |
| GEPABackbone=GPT-5, Number of LLM calls=502026.02 | 43.04 | 33.68 | |
| GPT-5 (single call)Backbone=GPT-5, Number of LLM calls=12026.02 | 20.64 | 0 |